🐾

Peter's Weight

Puppy growth tracker β˜… Born Dec 15, 2025 β˜… Embark genetic profile
Pete's weigh-ins
Data fit (~ lb)
Projected
Breed prediction (~ lb)
Breed range
First weigh-in
Current
Data fit (adult)
Breed-weighted
Curves are 3-parameter Gompertz W(t) = AΒ·exp(βˆ’exp(βˆ’(tβˆ’c)/b)) with b=12.7 wk, c=14.1 wk, calibrated to a 28%/72% blend of Salt 2017 size category IV/V centiles. See the Analysis tab for math and sources.

This tab derives a peer-reviewed adult-weight prediction for Peter from his confirmed Embark profile, and compares it to a separate prediction fit purely to his weigh-in trajectory.

1Confirmed genetic profile

Embark groups Medium and Standard Poodles together (and Toy and Miniature together) because each pair is genetically the same breed, distinguished in continental European registries (FCI No. 172) by height only. The profile sums to 100% β€” no unresolved bucket.

BreedAdmixtureSalt 2017 size
Golden Retriever57%Cat V (30–40 kg)
Labrador Retriever15%Cat V
Siberian Husky11%Cat IV (15–30 kg)
Bulldog4%Cat IV
Poodle (Medium & Standard)4%Cat III/IV
Saint Bernard3%Cat VI (>40 kg)
Poodle (Toy & Miniature)2%Cat I (<6.5 kg)
Chinese Shar-Pei1%Cat IV
Collie1%Cat V
German Shepherd1%Cat V
Australian Shepherd1%Cat IV/V

Two of the small percentages carry disproportionate genetic information: 2% Toy/Mini Poodle sits at the opposite end of the size spectrum from the bulk of the profile, and 3% Saint Bernard brings genuine giant-breed alleles. Their effects on the linear-mean prediction roughly cancel (~+4.8 lb from Saint Bernard, ~βˆ’1.1 lb from Toy/Mini relative to ancestry-mean assumption), but their effects on the underlying genotype distribution do not.

2Per-breed adult weight (male)

Peter is male, so all references are male-specific. For each breed I record the AKC/FCI/KC standard, reconcile against measured-population data where available, then settle on a point estimate and a low/high.

2.1 Golden Retriever (57%)

SourceMale range
AKC standard (GRCA, 1990)65–75 lb (29.5–34 kg)
FCI Standard No. 11129–34 kg
The Kennel Club (UK)weight not specified
Morris Animal Foundation GRLS (n=3,044)not published as median; 37% of cohort overweight/obese

Synthesized: 70 lb (low 65, high 75). All registries that publish a weight converge on the same range; no measured-population mean has been published from GRLS.

2.2 Labrador Retriever (15%)

SourceMale value
AKC standard ("in working condition")65–80 lb (29.5–36.3 kg)
FCI Standard No. 12229–36 kg
Helmink 2000 (J Anim Sci, n=880, Gompertz Wmax)31.4 kg = 69.2 lb
O'Neill 2018 (PLOS ONE, VetCompass UK, n=12,069 males)mean 35.2 kg = 77.6 lb (SD 13 lb)
Penell 2019 (Acta Vet Scand, n=12 males, 6–12 yr)28.7 kg β†’ 32.1 kg

The 8-lb gap between Helmink's working-line dogs and the VetCompass pet-population mean is the well-documented Lab obesity tendency (Raffan 2016 Cell Metab: ~25% carry the POMC 14-bp deletion adding ~4 lb per allele).

Synthesized: 73 lb (low 65, high 80). Splits the difference between working-line asymptote and pet-population mean.

2.3 Siberian Husky (11%)

Synthesized: 52 lb (low 45, high 60). All four registries (AKC 45–60, FCI 20.5–28 kg, KC 20–27 kg, UKC 45–60) align within 1 kg.

2.4 Bulldog (4%)

Synthesized: 55 lb (low 50, high 60). Compromise between AKC ideal (~50 lb) and measured pet-population mean (~60 lb, O'Neill 2019). Largest standard-vs-measured gap of any breed reviewed; reflects breed obesity, not genetic body-size potential.

2.5 Poodle (Medium & Standard combined, 4%)

Synthesized: 50 lb (low 35, high 70). Wide range reflects genuine genetic-test ambiguity. AKC descriptive: Standard male 60–70 lb; Embark Standard Poodle: male 50–70 lb; Wisdom Panel "Med + Std": 15–69 lb.

2.6 Saint Bernard (3%) and Toy/Miniature Poodle (2%)

Saint Bernard β€” AKC: males 140–180 lb (63.5–81.6 kg). FCI: minimum 70 cm at withers. Synthesized: 160 lb (low 140, high 180).

Toy + Miniature Poodle β€” combined envelope ~4–15 lb. Synthesized: 10 lb (low 5, high 15).

2.7 Smaller-percentage breeds

BreedPctMid (lb)LowHigh
Chinese Shar-Pei0.01524560
Collie0.01676075
German Shepherd0.01776590
Australian Shepherd0.01575065

3Growth model

3.1 Why Gompertz, not Logistic

TjΓΈrve & TjΓΈrve 2017 review canine growth modeling. The logistic model is symmetric (inflection at 50% of asymptote). Gompertz inflects at 1/e β‰ˆ 36.8% β€” asymmetric, with fast acceleration early and a long tail. Puppies hit half-adult-weight well before age-mid, so Gompertz is the better mechanistic fit.

W(t) = A Β· exp(βˆ’exp(βˆ’(t βˆ’ c) / b)) A = adult weight, c = inflection age, b = scale parameter (all in weeks)

3.2 Calibrating Gompertz to Salt 2017 size categories

Peter's breed-weighted adult lands at 67 lb / 30.4 kg β€” right at the IV/V boundary. About 72% of his admixture is in Cat V (Golden + Lab); 15% in Cat IV (Husky + Bulldog); 4% Poodle straddles III/IV; 9% spread elsewhere.

Fitting Gompertz W/A = exp(βˆ’exp(βˆ’(t βˆ’ c)/b)) to the Pete-weighted column at three anchor weeks (8, 16, 26) yields:

  • b = 12.7 weeks (scale parameter)
  • c = 14.1 weeks (inflection age)
Verification.
Week 8: exp(βˆ’exp(βˆ’(8βˆ’14.1)/12.7)) = 0.199 β†’ 19.9% βœ“
Week 52: exp(βˆ’0.052) = 0.949 β†’ 95% βœ“

4Breed-weighted adult weight prediction

4.1 Point estimate

A_breed = 0.57Β·70 + 0.15Β·73 + 0.11Β·52 + 0.04Β·55 + 0.04Β·50
        + 0.03Β·160 + 0.02Β·10 + 0.01Β·52 + 0.01Β·67 + 0.01Β·77 + 0.01Β·57
        = 68.30 lb β‰ˆ 68 lb

4.2 Breed-standard range (band)

Breed-standard range: 62–75 lb. Envelope if Peter sits at the low or high end of every breed's standard simultaneously.

4.3 Genetic prediction uncertainty

Plassais et al. 2019 (Nat Commun) attribute ~95% of standard-breed-weight variance to ~14 genes. Arvelius et al. 2023 measured hΒ² = 0.51 for adult body weight β€” meaning genetics caps prediction rΒ² at ~0.5; the other half is environment, BCS, and developmental factors.

Peter: point estimate 68 lb; 1Οƒ β‰ˆ 60–76 lb; 95% genetic CI β‰ˆ 53–83 lb.

5Data-fit adult weight prediction

The same Pete-weighted Gompertz curve, but solving for A from each weigh-in (back-out adult weight from current weight given the curve shape), then taking a recency-weighted average. Quadratic weights (iΒ² for the i-th observation) give heavier weight to later, more-predictive points.

Data-fit point estimate: ~55 lb. Implied A is rising over time as Peter's growth pulls the recency-weighted estimate upward.

Comparison

MethodAdult (lb)
Breed-weighted (genetic)68
Data-fit (Pete-weighted Gompertz)55

The 13-lb gap is well within the genetic CI for a mixed-breed prediction (~Β±20–25%). Embark itself is candid that single-dog mixed-breed predictions can miss by ~25%. Final adult weight in the 58–62 lb range is the most defensible point estimate combining genetics with his actual trajectory.

6Sources

Growth modeling

  • Salt 2017 β€” Growth standard charts for monitoring bodyweight in dogs of different sizes. PLoS ONE 12(9):e0182064. PMC5584974
  • Helmink 2000 β€” Breed and sex differences in growth curves for two breeds of dog guides. J Anim Sci 78(1):27. PMID 10682799
  • TjΓΈrve 2017 β€” Use of Gompertz models in growth analyses. PLoS ONE 12(6):e0178691. PMC5459448

Genetic basis of body size

  • Sutter 2007 β€” A single IGF1 allele is a major determinant of small size in dogs. Science 316:112. PMC2789551
  • Plassais 2019 β€” Whole-genome sequencing reveals variants influencing morphology. Nat Commun 10:1489. DOI
  • Rimbault 2013 β€” Derived variants at six genes explain nearly half of size reduction. Genome Res 23:1985. PMC3847769

Heritability and population variation

  • Arvelius 2023 β€” Heritability and genetic trend of body weight in dogs. J Anim Sci 101:skad173. DOI
  • Salt 2023 β€” Exploration of body weight in 115,000 young adult dogs of 72 breeds. Sci Rep 13:22. DOI

Genetic-test methodology